Search Constraints

Start Over
Filtering by: Thesis Degree Discipline Engineering, Electrical and Computer Remove constraint Thesis Degree Discipline: Engineering, Electrical and Computer

Search Results

  • Modular Verification of Hierarchical Component-Based Software Systems

    Resource Type:
    Thesis
    Creator:
    Baak, James Alec William
    Abstract:
    Using Component-based Software Engineering approaches with Formal Methods has seen an influx of interest in the recent decades. The joining of these two disciplines have been stifled though due to unclear component specifications and expensive formal verification techniques, which hurt the reusability and scalability of complex software systems. In this work, we expand on current component-port-connector metamodels for formally specifying a system's architectural and behavioural requirements into a hierarchical component system structure by using abstract Composite Components. The Composite Components of a system model can then utilize modular verification for isolating the verification process into modules surrounding Composite Components and generating higher level properties. We formalize our metamodel in Alloy 6 and present a template for specifying system properties for modular verification which enables the reuse of previous verification efforts on satisfied modules. We conclude with an example case study system and analysis of the modular verification strategy.
    Thesis Degree:
    Master of Applied Science (M.App.Sc.)
    Thesis Degree Discipline:
    Engineering, Electrical and Computer
    Date Created:
    2023

  • A Web Based Architecture to Operationalize Geospatial Simulation Environments

    Resource Type:
    Thesis
    Creator:
    St-Aubin, Bruno
    Abstract:
    Simulation is inherently multi-disciplinary. It requires knowledge about the system under study, expertise in simulation theory to define models and programming skills to implement models. Geospatial simulation requires an additional layer of expertise in topology, geospatial data structures, spatial analysis, computational geometry, and other related topics. Commercial modeling and simulation software can be used to provide an environment to facilitate simulation studies for users. However, these software tend to be narrowly scoped to specific business applications and tightly couple model and simulator. As such, it is difficult to expand their usage and reuse them outside of the application domain they were intended for. The Discrete Event System Specification (DEVS) is a modular and hierarchical simulation formalism that clearly separates the model, simulator and experiments. It can be used break down the disciplinary silos within which single-use simulators are built and allow users to study real-world systems from a broad range of application domains. In this research, we present an architecture that facilitates the operationalization of DEVS based, geospatial simulation environments in multidisciplinary projects. The architecture relies on a clear definition of roles and responsibilities to leverage the different skillsets in an organization. It considers a series of business processes for modelers, subject matter experts, web developers and end users. It relies on a web-based architecture to provide simulation as a service capability and support users across the entire simulation lifecycle. It seeks to democratize DEVS simulation by making use of the strengths and skills available in larger organizations and by providing the necessary tools for collaboration. Importantly, it preserves key features of DEVS (genericity, modularity, flexibility, etc.) and encourages users to follow best practices in model documentation to foster model reusability and improve model discoverability. It relies on modeling and simulation as a service to overcome technological barriers of entry for DEVS simulation and provide a set of reusable tools to design simulation-based, web applications for end users.
    Thesis Degree:
    Doctor of Philosophy (Ph.D.)
    Thesis Degree Discipline:
    Engineering, Electrical and Computer
    Date Created:
    2023

  • QoS Performance Measurement Through SRv6 Network Programming for Smart City IoT Traffic

    Resource Type:
    Thesis
    Creator:
    Kuri, Sajib Kumar
    Abstract:
    Segment Routing over IPv6, also known as SRv6, is a modern networking solution that aims to improve the current Internet of Things (IoT) network's reliability, availability, and scalability. Performance measures are required to evaluate SRv6 behaviors or functions. The proposed work aims to provide real IoT traffic profiles to assess the performance of SRv6 behaviors. In particular, a three-module SRv6 programming model is proposed to measure the performance of SRv6 policy headend and endpoint behavior and ensure reliability and quality of service (QoS). Moreover, a novel finder algorithm for maximum receive rate (MRR) benchmarking is proposed, which can outperform existing techniques in terms of throughput/bandwidth performance while maintaining the same computational resources. Finally, implementation results provide insights into forwarding different IoT use-cases traffic based on the functional service requirements. That also ensures a higher usage level of existing IoT networks, minimizing the need for additional capacity and lowering network costs.
    Thesis Degree:
    Master of Applied Science (M.App.Sc.)
    Thesis Degree Discipline:
    Engineering, Electrical and Computer
    Date Created:
    2022

  • Optimal Selection of IPsec-Based Security Mechanisms in Resource Constrained IoT Environment

    Resource Type:
    Thesis
    Creator:
    Mahmoud, Abdullah Fahmi Mustafa
    Abstract:
    Internet of Things (IoT) is a modern-day technology that supports many different applications such as smart cities, e-health, and smart homes. However, the specific IoT nature of connecting various heterogeneous devices complicates the implementation of conventional security mechanisms. In this work, a device-based security approach has been proposed to assign optimal security mechanisms to the set of heterogeneous IoT devices based on their available resources and the system requirements. To achieve the proposed approach, the security overhead equation was formulated to include 3 parameters: RAM usage, energy consumption and throughput. A hardware implementation was used to measure these parameters and to calculate the security overhead for the tested security mechanisms. The Pareto frontline was used to select the optimal security mechanism that minimizes the security overhead per device while maximizing the system requirement. The selection algorithm was tested in a simulation of 50 heterogeneous devices that ran 30 security mechanisms.
    Thesis Degree:
    Master of Applied Science (M.App.Sc.)
    Thesis Degree Discipline:
    Engineering, Electrical and Computer
    Date Created:
    2023

  • Knowledge Graph Generation for Unstructured Data Using Data Processing Pipeline

    Resource Type:
    Thesis
    Creator:
    Sukumar, Sushmi Thushara
    Abstract:
    Research on effective usage of Machine Learning (ML) and Natural Language Processing (NLP) techniques are taken up to mitigate the problem of extracting information from huge volumes of unstructured data available on the Internet without losing valuable information. Constructing Knowledge Graph is one such application to query and extract unstructured data. The data is passed through a coreference resolution module using Neuralcoref, a named entity linking module using Wikifier API, and a relationship extraction module using two models, namely, OpenNRE and REBEL, and stores the results as a KG in Neo4j with its corresponding entities and relationships. Experiments were conducted on an unstructured dataset (BBC news dataset) containing text data to analyze the results obtained from the pipeline. The results obtained in the relationship extraction stage were analyzed for evaluation purposes and achieved 61.4% and 87% accuracy through the OpenNRE and REBEL models, respectively.
    Thesis Degree:
    Master of Applied Science (M.App.Sc.)
    Thesis Degree Discipline:
    Engineering, Electrical and Computer
    Date Created:
    2023

  • Tilted Fibre Bragg Grating Based Encoder System for Rapid Temperature Measurements

    Resource Type:
    Thesis
    Creator:
    Smith, Eric William
    Abstract:
    This master's thesis presents a tilted fibre Bragg grating (TFBG) as a form of optical encoder system to measure temperature through single wavelength interrogation. The TFBG is interrogated with a single wavelength source that is positioned over a cladding mode resonance, when the temperature shift occurs, the shift temperature modulates the transmission of the device as the TFBG spectrum shifts and the comb-like structure of the cladding mode resonances pass over the interrogation wavelength. Through this modulation of the transmission, the speed and magnitude of the temperature shift can be obtained. The presented TFBG encoder system demonstrates a higher sensitivity to temperature relative to thermocoules and rise times as low as 500μs.
    Thesis Degree:
    Master of Applied Science (M.App.Sc.)
    Thesis Degree Discipline:
    Engineering, Electrical and Computer
    Date Created:
    2023

  • System Analysis for Satellite Networks with Free-Space Optical Links

    Resource Type:
    Thesis
    Creator:
    Liang, Jintao
    Abstract:
    In this thesis, we analyze several system parameters and performances for free-space optical satellite networks (FSOSNs), including phasing parameter, optical satellite link budget, tradeoff between network latency and satellite transmission power, and network latency minimization based on satellite transmission power constraints. We investigate the phasing parameter for Starlink Phase 1 Version 3 and Kuiper Shell 2 constellations. We find the best value of the phasing parameter. We investigate the link budget for laser inter-satellite link (LISL) and uplink/downlink and the effect of many constellation parameters on the satellite transmission power. We examine the tradeoff between network latency and satellite transmission power in FSOSNs from Starlink Phase 1 Version 3 and Kuiper Shell 2 constellations for different LISL ranges and inter-continental connections. We investigate the minimization of total network latency in the FSOSN resulting from Starlink Phase 1 Version 3 constellation for different LISL ranges and satellite transmission power constraints.
    Thesis Degree:
    Master of Applied Science (M.App.Sc.)
    Thesis Degree Discipline:
    Engineering, Electrical and Computer
    Date Created:
    2023

  • Supporting End Users in Securing IoT-Enabled Smart Home Devices.

    Resource Type:
    Thesis
    Creator:
    Sereda, Bohdana
    Abstract:
    Smart home devices have gained popularity among users and become a usual thing found in one's home. While they provide comfort and automation, they also pose various security risks, which can be addressed in up-to-date security documentation. Besides, it is important to acknowledge users' responsibility for their device's security and to provide them with helpful resources. In this thesis, we overview several Internet of Things (IoT) security documents and analyze their content to understand the role of users in each of them. Having a trustworthy resource on smart home security for the users is essential to providing security on their end. Therefore, based on our findings from the analysis, we present a prototype tool that provides security recommendations to smart home users. After that, we evaluate the tool during a user study, which suggests that our target audience would, in fact, benefit from having accessible security recommendations and tools.
    Thesis Degree:
    Master of Applied Science (M.App.Sc.)
    Thesis Degree Discipline:
    Engineering, Electrical and Computer
    Date Created:
    2022

  • Silicon-Based Polarization-Insensitive Optical Antenna Design and Experimental Characterization of Optical Phased Arrays

    Resource Type:
    Thesis
    Creator:
    Xin, Xiaochen
    Abstract:
    Optical Phased Array (OPA) has emerged as one of the most popular technologies in recent years. The integration of optoelectronics components on-chip allows the OPA to steer the beam to achieve ranging, detection, and free-space communication without having any moving parts. The thesis includes two parts. Firstly, a polarization-independent optical surface grating antenna designed for OPA is presented. The designed antenna emits both quasi-transverse electric (TE) and quasi-transverse magnetic (TM) modes towards the same angle with similar beamwidth. With the increasing demanding for mode-division multiplexing systems, the incorporation of such antenna in an OPA system allows an additional channel of data transmission while preserving the steerability of the array. In the second part of the thesis, the optical testing setup for a fabricated on-chip OPA system is designed and presented. With the testing setup designed and assembled, a comparison between the observed and simulated far-field images is also presented.
    Thesis Degree:
    Master of Applied Science (M.App.Sc.)
    Thesis Degree Discipline:
    Engineering, Electrical and Computer
    Date Created:
    2023

  • TC-GVF: Tensor Core GPU Based Vector Fitting Via Accelerated Tall-Skinny QR Solvers

    Resource Type:
    Thesis
    Creator:
    Kukutla, Vinay Ven
    Abstract:
    In this MASc thesis work, the widely used system identification method, Vector Fitting (VF), is advanced for adoption in the emerging GPU architectures with Tensor Cores. Since the VF algorithm is iterative in nature, improving its computational cost and parallel efficiency on mixed CPU and GPU environments is critical in reducing the overall time needed for convergence. Algorithmic advancements are introduced to provide significant speedups to the most computationally expensive steps in the VF process, QR factorization and the solution to a set of linear equations. Furthermore, Nvidia's new Tensor Core architecture is leveraged to provide further performance improvements. The application examples for modeling of large multiport models with high-speed modules demonstrated orders of magnitude speed-up compared to the existing work in the literature.
    Thesis Degree:
    Master of Applied Science (M.App.Sc.)
    Thesis Degree Discipline:
    Engineering, Electrical and Computer
    Date Created:
    2023